The results of a study are often presented toward the middle-to-end of a report. This section should include the results of the data analysis (i.e., the statistics for quantitative research and the categories for qualitative research). When areas of the results section are missing or confusing, it can be difficult for you to understand what the study found.<\/p>
When reading a report, you should never feel confused about how the authors arrived at a particular number or result. The numbers should add up correctly in both text descriptions and data visualizations. When report authors are sharing numbers based on calculations, those calculations should be described for you to assess their validity (appropriateness) in the context of the overall report. When looking at visual representations of the data (i.e., charts and graphs), they should be labeled so that it is easy to understand which numbers are represented in the figures.
Whenever data are being described, the total number of respondents or responses for all participants (or \u201cN\u201d) or for subgroups of participants (or \u201cn\u201d) should be clearly indicated. Percentages should include the context of the total number of participants. Similarly, the total N that is being broken down in any form of graph, table or chart should be labeled. Including Ns are important for the interpretation of the data. For example, if the results show 80 percent, it is important to know how many individuals this represents.<\/p>\n
When the N and n are missing from a report, it can be difficult for you to evaluate whether there were enough participants to draw conclusions from the results. Additionally, it is important for you to see how many participants were in subgroups so you can further understand the sample and the results.
Sometimes, study participants choose to skip a question or they do not complete the entirety of a survey. Thus, a study result may contain missing data. Reports should include information about this missing data. This can be done when the sample is described; for example, missing data may result in a participant\u2019s data being excluded from the analysis. Missing data can also be described at various points throughout the report when study results are shared (e.g., by noting missing data as a category in a graph or table).<\/p>\n
When information about missing data is not included, it can be difficult for you to understand the numbers (see point a above) and the N and n (see point b above). Reporting missing data helps you easily see how numbers add up correctly and how many participants were included in each analysis.
When results are presented in a report, you should be able to discern how the researchers arrived at the result and how the result fits in with the study\u2019s purpose. For example, did the result answer a research question identified earlier in the paper? Was the data collection and analysis for the result described in the methodology section? If a finding is being presented and it is unclear how the finding resulted from the study\u2019s purpose, research questions and methodology, then the result is without context. This lack of context can make it difficult for you to ascertain the finding\u2019s importance to the report overall. For example, if you see a result in a report that doesn\u2019t seem related to the study\u2019s purpose, it is difficult to understand why the result is included and how\/why it contributes to the goals of the study.
Good data visualization<\/span><\/span> will not only visually present the data but it will also add something to one\u2019s understanding of the data. For example, a data visualization might represent trends or changes over time in the form of a line graph or show comparisons of different data points in a bar chart. Data visualizations can help you understand how results of the study relate to one another and relate to the study aims.<\/p>\n Data visualizations are quite diverse and certain methods of visualizing certain kinds of data can work better than others. Data visualizations are used to help explain data, so if they are confusing or hard to understand, then something is wrong. Data visualizations should be appropriate in size and scale so that the numbers represented are accurate. When reviewing data visualizations, you should also pay attention to whether the numbers add up correctly (e.g., a pie chart should add up to 100 percent). Also, pay attention to scale. When scale is not represented correctly, the relationship between data points can be exaggerated, distorted or just confusing. For example, if students could have scored between 0 and 100 on an exam, and the average scores are on a chart with numbers that range from 80 to 100, the differences between exam scores are going to look larger than if the chart\u2019s numbers included the full range of 0 to 100 percent. <\/p>\nExamples<\/h2>\n
a. All numbers used in the report are easy to comprehend.<\/h3>\n
\n
b. An \u201cN\u201d is offered whenever data is being described in text or graph, table or chart.<\/h3>\n
\n
c. The report identifies missing data.<\/h3>\n
\n
d. It is clear where study findings fit in with the study\u2019s purpose, research question(s) and methodology.<\/h3>\n
\n
e. The data visualizations (graphs, charts and tables) enhance your understanding of the results.<\/h3>\n
\n
\n